Download The Future of Food: Scenarios for 2050

Published March, 2010
SYMPOSIA
The Future of Food: Scenarios for 2050
Bernard Hubert, Mark Rosegrant, Martinus A. J. S. van Boekel, and Rodomiro Ortiz*
ABSTRACT
This background article addresses key challenges of adequately feeding a population of
9 billion by 2050, while preserving the agroecosystems from which other services are also
expected. One of the scenario-buildings uses
the Agrimonde platform, which considers the
following steps: choosing the scenarios and
their underlying building principles, developing
quantitative scenarios, and building complete
scenarios by combining quantitative scenarios
with qualitative hypotheses. These scenarios
consider how food issues link to production, for
example, the percentage of animal vs. vegetal
calorie intake in the full diet. The first section of
this article discusses Agrimonde GO and Agrimonde 1 scenarios, which indicate that global
economic growth and ecological intensification remain as main challenges for feeding the
earth’s growing population toward the mid-21st
century. The second section provides the outcomes of the analysis of alternative futures for
agricultural supply and demand and food security to 2050, based on research done for the
International Assessment of Agricultural Science and Technology for Development. The last
section of this article provides a summary analysis of food systems and functions, as well as
the role of food technology that address some
of the global challenges affecting the supply of
more nutritious and healthy diets. It also highlights the food production by novel means (e.g.,
alternatives for animal products based on plant
materials) and increasing the presence of potentially health-promoting compounds in food to
improve human and animal health. Finally, this
article proposes priority areas that should be
included in further agri-food research.
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
B. Hubert, French Initiative for International Agricultural Research,
(FI4AR), Agropolis International, Ave. Agropolis, F-34394 Montpellier Cedex 5, France; M. Rosegrant, International Food Policy Research
Institute, 2033 K St, NW, Washington, DC 20006-1002; M.A.S.J. van
Boekel, Dep. Agrotechnology and Food Sciences, Wageningen Univ.,
PO Box 8129, 6700 EV Wageningen, The Netherlands; R. Ortiz, Centro Internacional de Mejoramiento de Maíz y Trigo (CIMMYT). R.
Ortiz, current address: Martin Napanga 253, Apt. 101, Miraflores, Lima
18, Peru. B. Hubert, M. Rosegrant, and M.A.S.J. van Boekel contributed equally to this manuscript. Received 21 Sept. 2009. *Corresponding author ([email protected]).
Abbreviations: AKST, Agricultural Knowledge Science and
Technology; FAO, Food and Agriculture Organization of the United
Nations; GDP, gross domestic product; IAASTD, International
Assessment of Agricultural Science and Technology for Development;
IMPACT, International Model for Policy Analysis of Agricultural
Commodities and Trade; MEA, Millennium Ecosystem Assessment;
OECD, Organisation for Economic Co-operation and Development.
P
roviding humankind with enough food has been a challenge
throughout the ages. This topic remains of importance, but
food production has changed considerably in the last 50 to 100 yr.
Food security and quality improved tremendously in the industrialized world, but increasing obesity suggests humankind’s difficulty in handling overweight. The impact of food production
on the environment has also become problematic. However, on
a global scale, food security and quality are not yet realized, and
even though the situation has changed in the past century, we
are still faced with tremendous challenges that require new food
options to provide nutritious and healthy diets to overcome malnutrition. The following options may assist in this endeavor:
Published in Crop Sci. 50:S-33–S-50 (2010).
doi: 10.2135/cropsci2009.09.0530
Published online 6 Jan. 2010.
© Crop Science Society of America | 677 S. Segoe Rd., Madison, WI 53711 USA
All rights reserved. No part of this periodical may be reproduced or transmitted in any
form or by any means, electronic or mechanical, including photocopying, recording,
or any information storage and retrieval system, without permission in writing from
the publisher. Permission for printing and for reprinting the material contained herein
has been obtained by the publisher.
S-33
1. Increasing the content of micronutrients in the edible parts of crops through plant breeding (i.e., biofortification).
2. Producing protein-rich products by novel means,
based on plant materials, as alternatives for animal products.
3. Eliminating potentially toxic compounds in staple foods.
4. Reducing constitutive or microbial toxins in
selected staples that impact food quality, safety, and
human health.
5. Evaluating biofortification strategies in the context of other approaches, such as diversification of
diet, to improve the diets of nutritionally disadvantaged people.
World food prices rose dramatically between 2000
and 2008 before beginning to decline later in 2008. A
major cause of soaring food prices was the rapid growth in
demand for biofuels, which has diverted land from food
production. Other factors, many of them long term, have
also contributed to the current food supply-and-demand
situation. Rapid economic growth and urbanization, particularly in Asia, have driven rapid demand for meat and
for maize (Zea mays L.) and soybeans [Glycine max (L.)
Merr.] for livestock feed. Improved economic growth in
Africa has increased the demand for staples such as rice
(Oryza sativa L.) and wheat (Triticum aestivum L.). Meanwhile, agricultural productivity growth, especially in
developing countries, continues to drop and the decline
of global food stocks in the last 5 yr has led to very tense
cereal markets, worldwide. Growing water scarcity and
climate change are also increasingly affecting food production and prices. Poor and food-insecure households
are among the hardest hit by rising food prices and, subsequently, by the global economic recession. Although
households that are net sellers of food are benefiting, most
poor households are net buyers of food.
This article gives some scenarios of the future of food
toward 2050. Two scenarios are built on Agrimonde foresight models, which address challenges for feeding the
world (Agrimonde, 2009). The third scenario ensues from
analyzing alternative futures for agricultural supply and
demand, and food security. This article ends summarizing food systems and functions, and how food technology addresses some of these global challenges affecting the
supply of more nutritious and healthy diets.
AGRICULTURE AND FOOD IN THE
WORLD OF 2050: SCENARIOS AND
CHALLENGES FOR A SUSTAINABLE
DEVELOPMENT
Agrimonde was established as a collective instrument—led
by the French Initiative for International Agricultural
S-34
Research on behalf of the Institut National de la Recherche
Agronomique and the Centre de Coopération Internationale en Recherche Agronomique pour le Développement
(CIRAD)—for analyzing global food and agricultural
issues under the scenario of feeding 9 billion people by
2050, while preserving agro-ecosystems from which
other services and products are expected (including climate change, carbon storage, biodiversity, bio-energy, or
bio-materials) (Chaumet et al., 2009). The variables considered for the analysis are multifarious, including geopolitical, social, cultural, sanitary, economical, agronomical,
ecological, or technological, to cite just a few (Agrimonde,
2009). The global scale at which such issues are raised does
not preclude reflections at the regional level, which are
necessary to account for the diversity of the world’s food
and agriculture, and their interactions, especially through
trade that contains other key variables.
The classical scenario method is based on a first step
of exhaustive recording of all kinds of variables likely to
impact on the future of the system studied, within the
timeline chosen for a future’s study (De Jouvenel, 2000).
The classical method of scenario-building would not be
suitable considering the number and the diversity of variables, as well as the importance of articulating the regional
and global contexts. This exercise would have been both
unwieldy and largely indecipherable by combining the
hypotheses on all the key variables for the future of a given
agro-ecosystem investigated at both regional and global
levels. The method was, therefore, adapted by building
a tool based essentially on the complementarity of quantitative and qualitative analyses. The quantitative module Agribiom was developed by B. Dorin and T. Le Cotty
(CIRAD, Montpellier, France) by formulating quantitative hypotheses at the regional level, on a limited number of variables, thereby reducing the complexity, while
affording an entry point for in-depth qualitative reflection on all the dimensions of the agro-ecosystem. This
scenario-building considers the following main steps: (i)
choosing scenarios and their underlying building principles, (ii) developing quantitative scenarios, and (iii)
building complete scenarios by combining quantitative
scenarios with qualitative hypotheses.
Choice and Principles of the Scenarios
For the 2006–2008 phase, the Agrimonde project chose the
Millennium Ecosystem Assessment (MEA) scenarios, in
particular Global Orchestration, to analyze it from the
angle of food and agricultural systems, and to construct
a single new scenario that departed from those of the
MEA scenarios (Agrimonde, 2008). The MEA scenarios,
which are references in international debates, were originally built to study the future of ecosystems. Hence, they
are not necessarily the most relevant for considering the
future of food and agricultural systems. It is nevertheless
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
interesting to compare the two types of approaches: one
regarding ecosystems and the other regarding the human
activities that have the strongest impact on ecosystems.
As a baseline comparison, Agrimonde chose to reconstruct
the MEA scenario Global Orchestration, which is a trend
scenario on food consumption, but with different underlying
societal priorities. Global Orchestration is the MEA scenario
with the largest reduction of poverty and malnutrition. It is
based on both the liberalization of trade and on major technical advances in terms of agricultural yields. The priority
given to economic development in this scenario, nevertheless, results in an exclusively reactive management of ecosystems and environmental problems. This scenario was called
Agrimonde GO because it was reconstructed on the basis of
the quantification method adopted in Agrimonde, and because
the population hypotheses chosen for this scenario are not
precisely those used in the MEA.
The MEA scenarios are exploratory because they
explore the consequences of changing trends by starting
with the present situation. Some experts, including those
involved in the MEA, indicated the need for a desirable
scenario on the future of ecosystems. As a result, a new
scenario (Agrimonde 1) was developed. The hypothesis of
Agrimonde 1 uses as reference points a combination of the
MEA scenario and the one proposed by Griffon (2006),
who describes agriculture considering all characteristics
of sustainability and the potential and conditions of a
“doubly green revolution” (Conway, 1997). This type of
agriculture would be characterized by agricultural production technologies that both preserve ecosystems and
allow for development through agriculture in countries
lacking capital, where the implementation of production
systems requiring intensive use of equipment, pesticides,
and fertilizers is limited. The same “population pressure”
hypotheses are used for comparing the Agrimonde 1 scenario to the Agrimonde GO scenario.
Agrimonde 1 can be regarded as a normative forecasting scenario because it aims to explore the meaning and
conditions of existence of a scenario on the development
of a sustainable food and agricultural system. The idea was
to better understand the meaning of such development,
with the dilemmas and the main challenges that this type
of scenario entail, and through the changes and discontinuities that it implies.
The World in 2050, as described in Agrimonde 1, is
based above all on sustainable food conditions, allowing for
the reduction of inequalities in food and health through a
drastic reduction of both undernourishment and excessive
food intake. The World in 2050 will need to implement a
set of actions to intensify productive systems and to increase
production in most regions. These actions will meet the following objectives: satisfying the growing demand, allowing for the development of income from agriculture in rural
areas of the Global South, and developing environmentally
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
friendly agricultural practices. These two scenarios are
constructed differently; while Agrimonde GO is essentially
a trend scenario starting from the current situation, Agrimonde 1 is built on the basis of sustainability objectives that
are supposed to be met by 2050, and explores the trajectories that would enable them to be attained.
Two underlying principles constitute the Agrimonde 1
and Agrimonde GO scenarios:
1. Assessing the capacity for each large region of the
world to satisfy its food needs in 2050, thereby
implying that interregional trade would be considered only after evaluating the extent to which agricultural production in each region covered local
needs.
2. Identifying the effects of future population trends
independently of the large international migratory
flows, so that the implications of expected population explosions could be examined fully with regard
to each region’s capacity to feed its own population.
In its present form, Agrimonde 1 as a tool limits the
construction of scenarios for the world’s food and agriculture in 2050 in several ways. First, there are no precise and
complete quantitative estimations for the consequences of
climate change on the world’s agriculture. Consequently
climatic phenomena (greater variability, alterations in
rainfall, rising temperatures, or melting of certain areas)
have not been taken into account. Nonetheless, the panel
of experts, inspired by the scenarios from the Intergovernmental Panel on Climate Change, modulated their
hypotheses in relation to the surface areas under crops and
to the possible yields in 2050 in the different regions. Second, even if the notion of pressure on natural resources is
dominant in the analysis in various respects (e.g., deforestation resulting from the extension of farmlands, water
shortages induced by climatic and demographic changes,
or deterioration of the quality of the soil and water caused
by farming practices), the quantitative module does
not integrate indicators of the consumption of natural
resources, such as quantities of water or energy consumed.
Finally, Agrimonde 1 is based on the hypothesis that
agricultural development is a driving force of global economic development and poverty alleviation (World Bank,
2008). This tool nevertheless enables us to verify whether
the supposed regional increases in agricultural production
effectively contributes to sufficient economic development, especially to avoid mass migration.
Food Consumption in 2050
In the Agrimonde scenarios, as in the MEA scenarios, “food
availability” serves as an approximation of food consumption. It is calculated as the balance between the calorie
equivalent of quantities of available foodstuffs (production + imports − exports ± stock variations) to feed the
human population in a region (i.e., excluding animal
WWW.CROPS.ORG
S-35
feed, non–food uses, seeds, and postharvest losses), and
the number of inhabitants of that region. It reflects the
quantity of calories available to consumers, at home and
through other channels, and includes calories that will
be lost between the purchase of the products and their
ingestion. It should not be confused with the quantity of
calories actually ingested, which is difficult to estimate. In
terms of ingestion, the net energy needs of a human being
are around 2000 to 3000 kcal daily, depending on sex,
height, weight, and intensity of physical activity.
Food consumption trends are very different between
Agrimonde GO and Agrimonde 1 (Fig. 1). Agrimonde GO
uses the hypotheses from the MEA Global Orchestration
scenario in which economic growth largely explains consumption levels. Total availabilities at regional and world
levels are given in the MEA report, but they were not split
by product. Precise extrapolations were made in the Agrimonde report to quantify the food consumption hypotheses of the Agrimonde GO scenario. Agrimonde GO can
qualify as a trend scenario in terms of the evolution of the
total food calorie consumption, where economic growth
boosts consumption in all regions to reach a mean global
availability of 3590 kcal per capita daily and substantially
reducing undernourishment.
The Agrimonde 1 scenario is clearly distinguished from
the Agrimonde GO trend scenario. The income–food consumption nexus is not the main determinant because of
great concerns for health, equity, and the environment.
The hypothesis of food availability that the Agrimonde
expert panel selected for 2050 is 3000 kcal per capita daily
in all regions, notwithstanding certain regional particularities visible in the breakdown in terms of animal calorie
sources (monogastric, ruminants, and halieutic). This set
of hypotheses is in sharp contrast with the trends observed
between 1961 and the beginning of the 21st century. It corresponds to a slow growth of food availability per capita
in most regions up to 2050, except in sub-Saharan Africa,
where the per capita food availability will increase by 20%
in 50 yr, and the Organisation for Economic Co-operation and Development (OECD)–1990 region, where it will
decrease by one-fourth (Fig. 1). The 3000 kcal are broken
into 2500 kcal of plant products and 500 kcal of animal
products. Within animal products, the proportion due to
monogastrics is increasing in all regions, whereas the proportion due to ruminants is declining despite high levels in
OECD-1990 countries, the former Soviet Union and Latin
America, and an increase in sub-Saharan Africa. Calories
of aquatic origin increased their share in varying proportions, which are linked to regional productive possibilities.
Although the oceans are a considerable source of food production, fishing will face structural limits related to several
factors (overfishing, artificialization of the littoral, pollution, accelerated erosion of the biodiversity). It is assumed
that marine aquaculture can increase at a faster pace than
S-36
it has over the past 40 yr, but at a different pace depending
on the region. In Agrimonde 1, the pace of the development
of marine aquaculture is high in Asia, OECD-1990, and
Latin America, and moderate in the other regions. Relative
stability in per capita availability of calories from freshwater
fish is expected, as the existing (and increasing) tension over
freshwater availability prevents any increase in freshwater
fishing. Trends in relation to population increases in each
region were thereafter calculated.
The set of hypotheses on food consumption assumes
that people’s diets will depart from current tendencies as
they take into account the objectives of sustainable development, which will ensue from the mounting pressure
on resources and public health problems associated with
human diets. It is a very strong set of hypotheses, as it
implies that consumers, producers, and public policymakers will take into account the global and local impacts of
modes of food production and consumption on health and
the environment. This set of hypotheses corresponds to
four challenges:
1. The wide gap between the observed availability and the necessary availability for food security. The actual mean daily
availability in 2000 was close to 4000 kcal per capita
daily in the OECD-1990 zone and just under 4500
kcal per capita daily in the United States, whereas
the Food and Agriculture Organization (FAO) of
the United Nations deems satisfactory a mean daily
per capita availability of 3000 kcal to guarantee that
each individual has sufficient healthy food (FAO,
2002). These gaps can be explained by the distribution of diets within the population, by the fact
that in rich countries the 3000-kcal threshold may
be simply exceeded, and by a great proportion of loss
between the available food and actual consumption,
linked to consumption habits.
2. The importance of equity in a sustainable development
scenario. Instead of using the assumption suggested
by Collomb (1999) that each region attains at least
3000-kcal per capita daily threshold, with some
countries exceeding that level, Agrimonde chose to
test a stronger hypothesis that there will be a convergence of average availabilities of food worldwide.
3. The food/health nexus. A daily per capita availability of
3000 kcal may have positive consequences in terms
of public health by (i) maintaining the proportion
of undernourished people at a relatively low level,
thus reducing the risks of malnutrition in developing countries; and (ii) limiting overconsumption,
a source of nontransmissible food-related diseases
such as obesity. Public actions aimed at changing
food-related behaviors are a response to the current
increase in obesity.
4. The relationship between diet and the pressure on natural resources. The aim of adequately feeding 9 billion
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
Figure 1. Mean regional food availability (daily kcal per capita) trends in Agrimonde Global Orchestration (AGO) and Agrimonde 1 (AG1)
scenarios. The data used for this figure (1961–2003) ensued by reprocessing data from the Food and Agriculture Organization (FAO)
of the United Nations. FSU: former Soviet Union States, LAM: Latin America, OECD: Organisation for Economic Co-operation and
Development (or so-called developed world), MENA: Middle East and North Africa, SSA: sub-Saharan Africa.
people in 2050 implies that, irrespective of the production methods, there will be considerable pressure
on natural resources that will increase along with the
growing proportion of animal products in people’s
diets. The production of animal calories requires
a substantial volume of plant calories, water, and
energy. In addition, breeding ruminants generates
greenhouse gases directly or indirectly (e.g., through
animal fodder, processing, and transport). This last
component is increasing with the intensification of
production. Caution is nevertheless required, considering the environmental impact of animal production. One can also consider that there is an
advantage in producing animals that optimize the use
of plant resources (e.g., grazing on pastures, which
humans cannot digest). Systems have, however, been
intensified over the past 40 yr, which has resulted
in shrinking pastures and concentrates, especially for
grains. Producing ruminants still has the advantage
of using land that is often unfit for crops (e.g., high
altitudes, slopes, or semiarid areas), and of storing
carbon on such lands. Furthermore, ruminants also
have various uses because they represent a form of
capital for their owner, provide organic fertilizer, are
often used as draft animals, and are sources of food
and regular income for populations often among the
poorest in the world.
The World in 2050 in the Agrimonde Scenarios
The analysis of scenarios, in terms of coherence and action
levels, and their comparison, enabled the identification of
certain qualitative hypotheses in the Agrimonde 1 scenario.
On this basis, the factors that had not yet been considered
in the analysis, but that were likely to have a decisive impact
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
on the world’s food and agriculture during the period
leading up to 2050, were sought. These factors have been
grouped into seven main themes: (i) the global context, (ii)
international regulations, (iii) the dynamics of agricultural
production, (iv) the dynamics of biomass consumption,
(v) the actors’ strategies, (vi) knowledge and technologies
in the field of food and agriculture, and (vii) sustainable
development. A complete scenario was built by developing
hypotheses on these different dimensions, with a concern
for the overall coherence and plausibility of the scenarios.
A possible account of the Agrimonde 1 scenario is proposed
here, as well as that of Agrimonde GO, which corresponds
to the MEA experts’ forecasts (Carpenter et al., 2005). This
article will focus on Points vi and vii.
Agrimonde GO: Feeding the World by Making
Global Economic Growth a Priority
The global availability of calories for consumption as
food, per day and per capita, will increase by 818 calories between 2000 and 2050. The steepest increases will
be in Asia, sub-Saharan Africa, and Latin America, and
the number of children suffering from malnutrition in
developing countries will decrease by a factor of 2.5 during the first half of the century. This trend, stimulated
by the rapid economic growth and intense urbanization,
will be accompanied by a richer protein content of diets as
people consume more meat and fish. It will result in the
growth of obesity in many regions (Asia, Africa), where
new nutrition policies need to be implemented.
Technological development will allow for more
intensive farming, as well as for an extensive use of fertilizers and genetically modified crops. The vast majority of farms, both small and large, will become highly
mechanized and industrial. Local know-how will often
WWW.CROPS.ORG
S-37
be replaced by standardized industrial methods and the
variety of agricultural species will be reduced. Multinational fi rms are a predominant feature of this scenario, as
they will increase their share of plant and animal production, primarily through the development of new genetic
strains. Nevertheless, it needs to increase the cultivated
area by 18%, with yields close to 33,000 kcal ha−1 daily.
Agrimonde 1: Feeding the World
by Preserving Ecosystems
In 2050, diets in the various regions of the world would
converge regarding calorie intake. On average about 3000
kcal per capita daily would be available worldwide. Cultural particularities would nevertheless maintain some
diversity in the distribution of the various food sources.
Increasing income would not lead to a convergence of diets
toward western diets. Even though in certain regions, especially in sub-Saharan Africa, food consumption trends are
initially based on economic development, they also stem
from behavioral changes in most regions. For instance, in
a region like OECD-1990, the mean calorie consumption
has declined from 4000 to 3000 kcal per capita daily. This
steep downward trend is the result of less wastage by users
or in catering systems, and more effective nutrition policies. The maintenance of diversity of diets also helps to
solve problems of deficiencies in micronutrients, primarily through the consumption of fruit and vegetables. The
fast growth of the proportion of raw products compared
with processed products, recorded at the beginning of the
century, has leveled off. This is a symptom of the diversification of food systems. It also stems from regulations
that have placed strong constraints on agri-food companies’
information and communication on nutrition in the rich
countries, encouraging them to limit the degree of product
processing, while continuing to sell products that are innovative, in terms of practicality and variety.
From 2000 to 2050, the agri-industrial model, initially
clearly dominant, merges increasingly with the local food
and agricultural systems based on short circuits and on the
diversity of small and medium-sized farms and processing
enterprises, especially in the developing world. The tendency
toward standardization, internationalization, and concentration around a limited number of multinational firms declines.
This change is facilitated by national and regional strategies
to ensure food security, and by the considerable impact of
corporate social responsibility (CSR) on large firms’ strategies. The agri-food sector is strongly affected by CSR as
consumers in the rich countries prove to be more and more
concerned about food issues, due to the spread of the sustainable food concept and following the “hunger riots.” They
pressure agri-food firms, often via nongovernmental and
consumer organizations, to take on their particular role in
economic development and the reduction of malnutrition,
as well as in the struggle against obesity. According to this
S-38
scenario, the increase of cropping area needed is almost 39%
more than the current state, with yields varying from 20,000
to 30,000 kcal ha−1 daily. In this case, there is a huge need
for new models of agricultural activities facing new ways of
combining the ecological and productive functions of agroecosystems in the same area corresponding to a model that
can be qualified as “integrationist.” It is based on the combination of different types of productive systems in a given
territory, adapted to the local ecosystems in such a way as
to maintain it in the form of a mosaic of ecosystems producing a diversity of services (e.g., purifying and regulating
water resources, soil conservation, maintenance of landscape
structures and biodiversity, or carbon fixation). This model
involves different types of farming (such as livestock, forestry,
or crops) in the same territory, on the same farm or on different farms, overlapping to differing degrees (see the mode
of ecological intensification in the Agrimonde 1 scenario for
the North Africa–Middle East, sub-Saharan Africa, Latin
America, and Asia regions).
Ecological Intensification, Performance
Criteria, and (Ir)Reversibility of Choices
Today the concept of ecological intensification essentially
refers to tailoring technical options, rather than prescribing a set of processes that can be applied uniformly everywhere. These so-called technical options encompass social,
economic, spatial, and political options that are not incidental and have probably not been sufficiently explored.
However, enough is known about the options that have
accompanied the process of rationalization (also known
as “modernization”) of North American and European
agriculture, so that they enable us to clarify the conditions
required for a particular option.
In the Agrimonde 1 scenario, the agricultural performance criteria are no longer limited to tech-economic
indicators. They encompass a range of indicators at the territorial level that pertain to the efficiency of agricultural
practices regarding water quality, biodiversity, and soil
quality conservation, as much as on commercialized production. In this scheme, the different types of productive
systems described above are no longer exclusive, but they
are complementary to each other by allowing for efficient
management of the diversity of the ecosystems involved.
The Agrimonde 1 scenario is a fine illustration of such complementarity. For instance, in Latin America forests are
devoted no longer to clearing for land use or protection,
but to intermediate forms corresponding to various agroforestry models. In Asia humid areas are not all drained,
but rather they are valued as a source of grazing land in
dry seasons or for combined agricultural and aquaculture
projects. In North Africa–Middle East and in sub-Saharan
Africa, rangelands with low forage productivity become
key elements in grazing routes that use a diversity of environments and biological corridors, enabling the fauna and
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
flora to circulate. The same applies to hedges, small woods,
and orchards, habitats for many crop auxiliaries and coarse
substances that preserve the soil and low-lying vegetation from the effects of wind and rain. In the Agrimonde 1
scenario, farms with a low level of efficiency, in terms of
exclusively tech-economic criteria, play an important role
in this respect in 2050. They make the multifunctionality
of agriculture fully meaningful; that is, not only a farming
activity that provides goods and services apart from agricultural goods, whether for food or not, but also one of the
activities practiced in a territory by some of the households
living there. In this sense it is both the territory and the
households that are multifunctional, as agriculture as such
represents only one of these functions.
The Agrimonde 1 scenario integrates a change of viewpoint on the multifunctionality of agriculture, assessed as
essential by both the recommendations of the 2008 International Assessment of Agricultural Knowledge, Science and Technology
for Development (IAASTD, 2008) and by the World Development Report 2008 (World Bank, 2008) on agricultural issues.
One of the first tasks to make it meaningful would consist of
producing performance criteria to evaluate the accomplishment of the different functions, not merely to remunerate
them, but to frame them politically and to administer them.
It would become evident that in such a scheme the different types of agriculture complement one another rather than
having to fit into a single model (e.g., from commercial specialized to family multipurpose farming).
In both scenarios, the question remains as to the real
capacity for emerging new technology options, which are
affected also by other factors including social, economic, and
local development issues. It could prove difficult to break
away from past choices that are embedded in current technical solutions (e.g., mechanization, fertilizer and pesticide use,
or genetic engineering) as well as in cognitive systems (such
as knowledge and know-how, representations of nature, pollution, or landscapes) and in the values of the main actors
involved. Are we not trapped in a technical rationalization? It
is a sort of lock-in that other sectors have also experienced—
except we cannot do without agriculture!
FOOD SUPPLY AND DEMAND
FOR FOOD SECURITY
This section examines the new realities of the global food
system, presenting the outcomes of the analysis of alternative futures for agricultural supply and demand and food
security to 2050, based on research done for the IAASTD
(Rosegrant et al., 2009).
Drivers that Influence the Future of Food
Supply and Demand and Food Security
Drivers that influence the future of food supply and
demand and food security include any natural or humaninduced factors that directly or indirectly influence the
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
future of agriculture. Indirect drivers include demographic, economic, sociopolitical, scientific and technological, cultural and religious, and biogeophysical change.
Important direct drivers include changes in food consumption patterns, natural resource management, land
use, climate, energy, and labor. The key quantitative drivers in this scenario assessment are summarized below.
Baseline Quantitative Modeling Assumptions
The baseline case forecasts a world developing out to 2050
as it does today, without deliberate interventions requiring new or intensified policies. The key assumptions of
the reference case include:
1. Population. The baseline (as well as alternative policy
experiments) uses the United Nations medium variant projections (United Nations, 2005), with the
global population increasing from slightly more than
6.1 billion in 2000 to >8.2 billion in 2050. Population growth drives changes in food demand.
2. Overall economic growth. Economic growth assumptions are based on the TechnoGarden scenario of the
MEA (Carpenter et al., 2005). Incomes are expressed
as MER-based values. The TechnoGarden scenario
assumptions are near the midrange growth scenarios
in the literature for the world as a whole and most
regions. In some regions, such as sub-Saharan Africa,
the scenario is relatively optimistic.
3. Agricultural productivity. Agricultural productivity
values are based on the MEA (TechnoGarden scenario) and the recent FAO interim report projections
to 2030/2050 (FAO, 2006). The MEA assumptions
have been adjusted from the TechnoGarden scenario assumptions to conform to FAO projections
of total production and per capita consumption in
meats and cereals, and to our own expert assessment.
The main recent technological change developments, with continued slowing of growth overall,
have been taken into account. Growth in numbers
and slaughtered carcass weight of livestock has been
similarly adjusted.
4. Nonagricultural productivity. In the reference case,
in general, productivity growth is projected to be
lower in nonagricultural than in agricultural sectors.
The nonagricultural gross domestic product (GDP)
growth rates are based on the MEA TechnoGarden
scenario, but with adjustments to align with World
Bank medium-term projections. While the relatively
higher productivity in agriculture largely reflects the
declining trends in agricultural terms of trade, this is
not translated into higher output growth in agricultural sectors relative to nonagricultural sectors. This
broadly confirms Engel’s Law that the budget share
of food falls with increasing income.
WWW.CROPS.ORG
S-39
5. Disparities in growth rates among developing countries are projected to remain high, while more
developed regions will see more stable growth out to
2050. Developed regions will see relatively low and
stable to declining growth rates between 1 and 4%
yr−1. Latin America is also expected to experience
stable growth rates, though slightly higher than for
developed regions—between 3.5 and 4.5% yr−1. The
GDP growth in East and Southeast Asia is expected
to be stable, with relatively high rates of 4 to 7%
yr−1. In particular, China’s economy is projected
to slow from the 10% growth in recent years to a
more stable rate of 5.6% yr−1. On the other hand,
growth in South Asia—following strong reforms
and initiatives focusing on macroeconomic stabilization and market reforms—is expected to lead to
improved income growth in that subregion of 6.5%
yr−1. The Middle East and North Africa is expected
to see GDP growth rates averaging 4% yr−1. Growth
in sub-Saharan Africa has been low in the recent
past, but there is room for recovery, which is projected to lead to modest to strong growth just under
4% yr−1. Growth in Central and Western Africa is
expected within the 5 to 6% range. Growth in East
and Southern Africa is expected at <4% out to 2025,
followed by more rapid growth of 6 to 9% by 2050.
6. Trade policies. Today’s trade conditions are presumed
to continue out to 2050. No trade liberalization or
reduction in sectoral protection is assumed for the
reference scenario.
7. Climate change. Climate change is both driving different outcomes of key variables of the baseline (such
as crop productivity and water availability) and is an
outcome of the agricultural projections of the reference run, due to land-use changes and agricultural
emissions, mainly from the livestock sector. Medium
energy outcomes are assumed in the baseline. The B2
scenario was used for the analysis. From the available
B2 scenario, the ensemble mean of the results of the
HadCM3 model for the B2 scenario was used. The
pattern scaling method applied was that of the Climate Research Unit, University of East Anglia. The
“SRES B2 HadCM3” climate scenario is a transient
scenario depicting gradually evolving global climate
from 2000 through 2100.
8. Biofuels. The baseline, based on actual national biofuel plans, assumes continued expansion in production of biofuels through 2025, although the rate of
expansion declines after 2010 for the early rapid
growth countries such as the United States and Brazil. Under this scenario, significant increases in biofuel feedstock demand occur in many countries for
commodities such as maize, wheat, cassava (Manihot esculenta Cranz), sugar, and oil seeds. By 2020,
S-40
the United States is projected to put 130 million t
of maize into biofuel production; European countries will use 10.7 million t of wheat and 14.5 million t of oil seeds; and Brazil will use 9 million t
of sugar equivalent. We hold the volume of biofuel
feedstock demand constant starting in 2025 to represent relaxed demand for food-based feedstock crops
created by the rise of new technologies that convert
nonfood grasses and forest products.
Models Used in the Study
Two types of models were used for the study: partial
agricultural equilibrium models and computable general
equilibrium (CGE) models. Both types were used for
analyses at the national (India and China) and regional or
global levels. The partial equilibrium agricultural sector
model—International Model for Policy Analysis of Agricultural Commodities and Trade, or IMPACT (Rosegrant
et al., 2002)—provided insights into long-term changes in
food demand and supply at a regional level, taking into
account changes in trade patterns using macroeconomic
assumptions as an exogenous input.
The IMPACT model was developed at the beginning
of the 1990s, on the realization that there was a lack of
long-term vision and consensus among policymakers and
researchers about the actions that are necessary to feed
the world in the future, reduce poverty, and protect the
natural resource base. This model has been used in several
important research publications, which examine the linkage between the production of key food commodities and
food demand and security at the national level. The most
comprehensive set of results for IMPACT are published
in the book Global Food Projections to 2020 (Rosegrant et
al., 2001). These projections are presented with details on
the demand system and other underlying data used in the
projections work, and cover both global and regionally
focused projections. This IMPACT model was further
expanded through inclusion of a water simulation model,
as water was perceived as one of the major constraints to
future food production and human well-being.
The Global Trade and Environmental Model
(GTEM)—A CGE model, developed by the Australian
Bureau of Agricultural and Resources Economics (Ahammad and Mi, 2005), was used to validate the GDP and population input data to achieve cross-sectoral consistency and
to implement trade analysis. The GTEM is a multiregion,
multisector, dynamic, general equilibrium model of the
global economy, which addresses policy issues with global
dimensions and issues where the interactions between sectors and between economies are significant. This includes
international climate change policy, international trade and
investment liberalization, and trends in global energy markets. In addition, the IAASTD analyses used the integrated
assessment model IMAGE 2.4 (Eickhout et al., 2006) for
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
climate change impacts and land use, and the livestock
spatial location–allocation model SLAM (Thornton et al.,
2002, 2006) for a more detailed livestock assessment.
Baseline Results
Food Supply and Demand
The baseline was a 3-yr average centered on 2000 for all
input parameters and assumptions for driving forces. Following this baseline, global cereal production increases
0.9% yr−1 for the 2000–2050 period. The year 2000 reflects
a 3-yr moving average for 1999 to 2001, and 2050 reflects
a 3-yr moving average of 2048 to 2050 unless noted otherwise. Growth of food demand for cereals slows during
the 2000–2025 period and again from 2025 to 2050, from
1.4 to 0.4% yr−1. The demand for meat products (beef,
sheep, goat, pork, and poultry) grows more rapidly but
also slows somewhat after 2025, from 1.8 to 1% annually.
Changes in cereal and meat consumption per capita
vary significantly among regions (Fig. 2 and 3). Over
the projections period, per capita demand for cereals as
food is expected to decline by 27 kg in East Asia and the
Pacific and by 11 kg in Latin America and the Caribbean.
On the other hand, demand is projected to increase by
21 kg in sub-Saharan Africa. Per capita meat demand is
projected to more than double in South Asia and subSaharan Africa, almost double in East Asia and Pacific,
and increase by 50% in the Middle East and North Africa.
In the developed countries, only a minor 4% increase is
projected, given that demand is already very high.
Total cereal demand is projected to grow by 1048 million metric t, or by 56%: 45% of this increase is expected
for maize; 26% for wheat; 8% for rice; and the remainder
for millet [Pennisetum glaucum (L.) R. Br.], sorghum [Sorghum bicolor (L.) Moench.], and other coarse grains. Rapid
growth in meat and milk demand in most of the developing world will put strong demand pressure on maize
and other coarse grains as feed. Globally, cereal demand as
feed increases by 430 million t for the 2000–2050 period;
that is, a 41% of total cereal demand increase. Slightly
more than 60% of total demand for maize will be used as
animal feed, and a further 16% for biofuels.
How will expanding food demand be met? For meat
in developing countries, increases in the number of animals slaughtered have accounted for 80 to 90% of production growth during the past decade. Although there
will be significant improvement in animal yields, growth
in numbers will continue to be the main source of production growth. In developed countries, the contribution
of yield to production growth has been greater than the
contribution of numbers growth for beef and pig meat,
while for poultry, numbers growth has accounted for
about two-thirds of production growth. In the future,
carcass weight growth will contribute an increasing share
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
of livestock production growth in developed countries, as
numbers expansion is expected to slow.
For the crops sector, water scarcity is expected to
increasingly constrain production, with little additional
water available for agriculture due to slow supply increases
and rapid shifts of water away from agriculture in key
water-scarce agricultural regions in China, India, plus
the Middle East and North Africa. Climate change will
increase heat and drought stress in many of the current
breadbaskets in China, India, and the United States, and
even more so in the already stressed areas of sub-Saharan
Africa. Once plants are weakened from abiotic stresses,
biotic stresses tend to set in, and the incidence of pest and
diseases tends to increase.
With declining availability of water and land that can
be profitably cultivated, area expansion is not expected
to contribute significantly to future production growth.
In the baseline, cereal harvested area expands from 660
million ha in 2000 to 694 million ha in 2020 before contracting to 632 million ha by 2050. The projected slow
growth in crop area places the burden to meet future
cereal demand on crop yield growth.
Although yield growth will vary considerably by commodity and country, in the aggregate and in most countries it will continue to slow. The global yield growth rate
for all cereals is expected to decline from 1.96% yr−1 in
1980–2000 to 1.01% in 2000–2050. In developed countries, annual average cereal yield growth is estimated at
0.96% yr−1 during 2000 to 2050, 0.9% in East Asia and
the Pacific, and 1.07% in South Asia. Slightly higher yield
growth is expected in the Middle East and North Africa,
Latin America and the Caribbean, and sub-Saharan
Africa; that is, at 1.16, 1.25, and 1.59% yr−1, respectively.
As can be seen in Fig. 4, area expansion is significant for
projected food production growth only in sub-Saharan
Africa (23%), Latin America and the Caribbean (9%), and
the Middle East and North Africa (7%).
Food Trade, Prices, and Security
In the last few years, real prices of food have increased
dramatically as a result of changes in biofuel/climate
policies, rising energy prices, declining food stocks, and
market speculation. Projections reported here show that
higher food price trends are likely to stay as a result of
increased pressures on land and water resources, adverse
impacts from climate variability and change, and rapidly
rising incomes in most of Asia. Given underinvestment in
agriculture over the past few decades, and projected slow
growth in investment in the baseline and poor government policies in response to rising food prices in many
countries, it is unlikely that the supply response will be
strong enough in the short to medium term.
Maize, soybean, rice, and wheat prices are projected to
increase by 60 to 97% in the baseline (Fig. 5) and prices for
WWW.CROPS.ORG
S-41
Figure 2. Per capita availability of cereals as food in 2000 and change for 2000 to 2050 by region. EAP: East Asia and the Pacific, LAC =
Latin America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA = sub-Saharan Africa.
Figure 3. Per capita availability of meats in 2000 and change for 2000 to 2050 by region. EAP: East Asia and the Pacific, LAC: Latin
America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.
Figure 4. Sources of cereal production growth (2000–2050) by region. EAP: East Asia and the Pacific, LAC: Latin America and the
Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.
Figure 5. International food prices ($US t−1) of selected grains in
2000, and projected for 2025 and 2050.
S-42
beef, pork, and poultry by 31 to 39%. Impacts of higher food
prices on net food purchasers will be substantial, depressing food demand in the longer term, increasing childhood
malnutrition rates, and reversing progress made in several
low-income countries on nutrition and food security.
World food trade is expected to continue to increase,
with cereals trade projected to increase from 257 million t
in 2000 to 584 million t by 2050, and trade in meat products rising from 16 million t to 64 million t. Expanding
trade will be driven by the increasing import demand from
the developing world, particularly sub-Saharan Africa,
East Asia and the Pacific, and South Asia, where net cereal
imports will grow by >200% (Fig. 6). Sub-Saharan Africa
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
Figure 6. Net trade in cereals in 2000 and projected for 2025 and 2050. EAP: East Asia and the Pacific, LAC: Latin America and the
Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.
will face the largest increase in food import bills despite
the significant area and yield growth expected during the
next 50 yr in the baseline. By 2050, the Middle East and
North Africa is expected to account for 33% of net cereal
imports, sub-Saharan Africa for 25%, and China for 19%.
With most developing countries unable to increase
food production rapidly enough to meet growing demand,
the major exporting countries—mostly in high-income
countries and in Eastern Europe and Central Asia—will
play an increasingly critical role in meeting global food
consumption needs. The United States and Europe are
a critical safety valve in providing relatively affordable
food to developing countries. However, given the strong
demand for food crops as feedstock for biofuels in the short
to medium term, net cereal exports in these countries are
projected to decline over the next decade before rebounding after food-crop use for biofuel feedstock is expected to
decline. For example, net maize exports from the United
States are expected to decline from 40 million t in 2000 to
17 million t in 2015 before rebounding and increasing to
62 million t by 2025. Net wheat exports are projected to
grow to 48 million t in Russia, 41 million t in the United
States, and to around 20 million t in Australia, Canada,
Central Europe, and Kazakhstan. Net meat exports are
expected to double in developed countries and to sharply
increase in Latin America. Brazil’s net meat exports are
expected to increase 10-fold over the 50-yr time horizon.
The substantial increase in food prices will slow
growth in calorie consumption due to both direct price
impacts and reductions in real incomes for poor consumers who spend a large share of their income on food. As a
result, there will be little improvement in food security for
the poor in many regions. In sub-Saharan Africa, daily calorie availability is expected to stagnate up to 2025 before
slowly increasing to 2762 kilocalories by 2050, compared
with 3000 or more calories available, on average, in most
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
other regions. Only South Asia (excluding India) fares
worse, with only 2654 kilocalories available on average by
2050. Several regions are projected to experience declining calorie availability between 2000 and 2025 (Fig. 7).
In the reference run, malnutrition among children up
to 60 mo will continue to decline slowly in most regions,
but remains high by 2050, with progress far below that
envisioned in the Millennium Development Goals (Fig.
8). Childhood malnutrition is projected to decline from
149 million children in 2000 to 130 million children by
2025 and 99 million children by 2050. The decline will
be greatest in Latin America at 51%, followed by Central/
West Asia and North Africa, and East Asia and the Pacific
at 46 and 44%, respectively. Progress is slowest in subSaharan Africa. By 2050, an 11% increase is expected—
to 33 million children in the region—despite significant
income growth and rapid area and yield gains, as well as
substantial progress in supporting services that influence
well-being outcomes, such as female secondary education
and access to clean drinking water.
ALTERNATIVE INVESTMENTS IN
AGRICULTURAL KNOWLEDGE,
SCIENCE AND TECHNOLOGY (AKST)
Three alternative AKST scenarios out to 2050 were
analyzed to examine their implications for food supply, demand, trade, and security. The first two scenarios
examine the outcome of different levels of investments
in crop yield and livestock numbers growth (AKST high
and AKST low). A third scenario analyzes the implications of even more aggressive growth in agricultural
R&D together with advances in complementary sectors
(AKST high plus). These include investments in irrigation infrastructure represented by accelerated growth
in irrigated area and efficiency of irrigation water use,
by accelerated growth in access to drinking water, and
WWW.CROPS.ORG
S-43
Figure 7. Calorie availability in 2000 and projected for 2025 and 2050. EAP: East Asia and the Pacific, LAC: Latin America and the
Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.
Figure 8. Number of malnourished children in 2000 and projected for 2025 and 2050, baseline. EAP: East Asia and the Pacific, LAC: Latin
America and the Caribbean, MENA: Middle East and North Africa, SA: South Asia, SSA: sub-Saharan Africa.
greater investments in secondary education for females, an
important indicator for human well-being (Table 1).
The AKST high variant that presumes increased
investment in AKST, results in higher food production
growth, which in turn reduces food prices and makes
food more affordable to the poor when compared with
the reference world. Under AKST high, cereal production increases by 7% and by an even stronger 17% under
the AKST high plus variant. Under AKST high, rice
prices decline by 46%, wheat prices by 57%, and maize
prices by 65%, compared with the 2050 baseline value.
On the other hand, if investments decrease faster than in
the recent past, prices would rapidly increase, by 96% for
rice, 174% for wheat, and 250% for maize compared with
the 2050 baseline value (Fig. 9).
Despite these strong changes in AKST behavior, yield
growth will continue to contribute most to future cereal
production growth under both the AKST low and AKST
high variants. Under AKST low, however, the contribution
of area growth to overall production growth is projected to
increase compared to the baseline, from 23 to 35% for subSaharan Africa, and from 11 to 29% in Latin America and the
Caribbean. For developing countries as a whole, area change
would contribute 13% to overall production growth, up
from a negative 4% (contraction of area) under the baseline.
S-44
This growth, coupled with rapid expansion of the livestock
population under AKST high, requires expansion of grazing
areas in sub-Saharan Africa and elsewhere, which could lead
to further forest conversion into agricultural use.
What are the implications of more aggressive production growth on food security and trade? Under AKST
high, developing countries cannot meet the rapid increases
in food demand through domestic production alone. As a
result, net cereal imports from developed countries would
increase by 70% compared with the reference run. Net
cereal imports are projected to increase from 72 to 125
million t in sub-Saharan Africa, and from 93 to 100 million t in the Middle East and North Africa, but drop by
almost half in China. Under AKST low, on the other
hand, high food prices lead to depressed global food markets and reduced global trade in agricultural commodities.
Sharp increases in international food prices as a result
of the AKST low variant depress demand for food and
reduce availability of calories. Average daily kilocalorie
availability per capita declines by 850 calories in subSaharan Africa, pushing the region below the generally
accepted minimum level of 2000 calories and thus also
below the levels of the base year 2000. On the other hand,
under the AKST high and AKST high plus scenarios,
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
Table 1. Assumptions for baseline and alternative Agricultural Knowledge, Science and Technology (AKST high, low) and infrastructure (AKST high plus) scenarios.
Parameter changes for
growth rates
Gross domestic product
growth
Livestock numbers growth
Food crop yield growth
Irrigated area growth
Rainfed area growth
Basin efficiency
Access to water
Base 2050
−1
3.06% yr
AKST high 2050
−1
3.31% yr
AKST low 2050
AKST high plus 2050
−1
3.31% yr−1
2.86% yr
Base numbers growth of ani- Increase in numbers growth
Reduction in numbers
Increase in numbers growth
mals slaughtered 2000–2050
by 20%
growth by 20%
by 30%
Livestock: 0.74% yr−1
Increase in animal yield by Reduction in animal yield by Increase in animal yield by 30%
Milk: 0.29% yr−1
20%
20%
Base yield growth rates 2000– Increase growth by 40%
Reduce growth by 40%
Increase growth by 60%
2050:
Cereals: 1.02% yr−1
Roots and tubers: 0.35% yr−1
Soybean: 0.36% yr−1
Vegetables: 0.80% yr−1
Fruits: 0.82% % yr−1
0.06
0.18
Increase by 25%
Decrease by 15%
Increase by 0.15 by 2050
Increase annual rate of
improvement by 50% relative
to baseline level
Increase overall improvement
by 50% relative to 2050 baseline level
Female secondary education
Figure 9. Cereal prices (US$ t−1) in 2050 per alternative Agricultural Knowledge, Science and Technology (AKST) scenarios.
calorie availability increases in all regions compared with
2000 and baseline levels.
Calorie availability—together with changes in complementary service sectors, including female secondary education, female-to-male life expectancy at birth, and access to
clean drinking water—can help explain changes in childhood malnutrition levels (Rosegrant et al., 2001). Under the
AKST high and AKST high plus variants, the number of
malnourished children in developing countries is projected
to decline by 24 and 56%, respectively, from 104 million
children under the baseline (Fig. 10). On the other hand,
if investments slow more rapidly and supporting services
degrade rapidly, then absolute childhood malnutrition levels could return to close to 2000 malnutrition levels at 137
million children in 2050 under the AKST low variation.
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
What are the implications for investment under these
alternative policy variants? Investment requirements for
developing countries in the baseline run for key sectors,
including public agricultural research, irrigation, rural
roads, education, and access to clean water, are calculated
at nearly US$32 billion per year at 2008 prices. As Fig. 11
shows, the much better outcomes in developing country
food security obtained under the AKST high plus variant
can be achieved at estimated annual investment increases
in the five key sectors of US$20 billion and are within
reach if the political will and resources are made available.
An Analysis of Food Production
The following trends are of importance in discussing the
future of food and the role of technology. An enormous
WWW.CROPS.ORG
S-45
Figure 10. Number of malnourished children, developing countries, projected 2000 to 2050 for alternative Agricultural Knowledge,
Science and Technology (AKST) scenarios.
Figure 11. Annual investment requirements (2000–2050) for agriculture and complementary service sectors using alternative Agricultural
Knowledge, Science and Technology (AKST) scenarios (US$ billion in 2008).
productivity increase has occurred in highly effective and
efficient farming systems. This improvement has resulted
in unprecedented food security in the industrialized world
but seems to lead to other problems as indicated below:
1. Industrialization of farming leads to alienation of
consumers toward their food.
2. Strong integration in food chains, also internationally, and consequently strong interdependencies.
3. Because of the abundant food supply (in the industrialized world), the role of the consumer has become
leading and necessary. In other words, food systems
are no longer supply driven but demand driven.
4. When food security is realized, food quality becomes
of importance, which materializes in a strong attention for the relation between food and health.
Food Systems
Food systems can be described as comprising four sets of
activities: producing food, processing food, packaging and distributing food, and retailing and consuming food (Ingram,
2008). The first activity of producing food is basically the
production of raw materials in agriculture, horticulture, animal husbandry, and aquatic production systems. The main
S-46
scientific disciplines involved are plant and animal breeding,
agronomy, soil science, water management, phytopathology,
and related disciplines. The main actors involved are farmers,
seed companies, fertilizer and pesticides industry.
The second activity is about the processing of raw
materials into food. This activity starts after harvesting and
four subactivities can be distinguished (Van Boekel, 1998):
stabilization, transformation, production of ingredients,
and production of fabricated foods. Stabilization implies
that measures are taken to prevent spoilage. The most
important cause of spoilage is microbial activity, which is
even dangerous as the microorganisms may be pathogenic
and are a threat to human health. This threat concerns the
very important aspect of food safety. Hence, many food
technology activities are directed toward the prevention,
or at least inhibition, of microbial growth. When microbial
growth is prevented, chemical and biochemical reactions
are the next cause of spoilage. This spoilage implies oxidation reactions and the so-called Maillard reaction, leading
to desired changes such as browning and flavor compounds,
and undesired changes such as loss of nutritive value and
toxicological suspect compounds. Biochemical changes
occur as a result of enzyme activity, which can lead to color,
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
flavor, taste, and texture problems. Transformation implies
that raw materials are changed into something different,
such as milk into cheese, wheat into bread, or barley into
beer. Production of ingredients is self-evident: sugar can be
extracted from sugar beets (Beta vulgaris L.) and sugarcane
(Saccharum officinarum L.), or protein and oil from soybeans.
This activity can also imply production with the help of
microorganisms or enzymes. Finally, production of fabricated foods implies that foods are composed or designed
from several raw materials (sauces, desserts, pastry are some
examples). Disciplines involved are food science and technology and nutrition, actors involved are artisanal as well as
industrial processors.
The third activity of packaging and distribution follows immediately after transformation. Some authors
would actually consider packaging to be part of processing. In any case, packaging is an essential element of
food technology as it protects the food against all kinds
of threats from the environment. These threats include
microorganisms, insects, water, oxygen, and physical
damage. Likewise, packaging technology can nowadays
be actively used to preserve foods by applying controlled
atmosphere and modified atmosphere. This implies that
the gas atmosphere influences metabolic reactions of the
food as well as of the microorganisms present in a desired
direction. Furthermore, packaging also functions as information carrier for the consumer, such as nutritive value,
presence of possible allergens, and any other relevant
information, including advertising. Disciplines involved
are food science, logistics, marketing, and actors are food
processors, middlemen, and retailers.
The last but not least activity is about retailing and consumption. Retailers have quite some power these days, as
they are able to influence the consumer directly by determining what to offer to consumers. They seem to have
appreciable governance in the food chain. Increasingly, globalizing markets are becoming important, as foods from all
over the world may end up at the consumer’s plate. While
this is not a bad development as such, this phenomenon
has several institutional implications, such as access to markets and a strong effect of rules and regulations that make
it sometimes difficult for developing countries to comply
with this development (Ruben et al., 2007). It also raises
the question whether or not such developments are not
enhancing sustainability problems. Actors involved are food
processors, retailers, and consumers, and governments, to
some extent, when regulation is involved.
Functions of Foods
Foods have many functions in society. First and foremost,
they supply energy and nutrients that humans need to live.
People need the macronutrients protein, fat, and carbohydrates. Next to that, the micronutrients vitamins and
minerals are essential. Furthermore, nonnutrients such as
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
fiber, antioxidants, and other bioactive components are
needed. However, not all proteins, fats, and carbohydrates
are the same. Generally, animal proteins (meat, milk,
eggs) are better from a nutritional point of view than plant
proteins. Fats also differ, and polyunsaturated fatty acids,
and especially the ω-3/6 fatty acids, are preferred over the
saturated ones. Carbohydrates that are absorbed in the gut
only supply energy to the body. Dietary fibers can also be
classified as complex carbohydrates that are not absorbed
but are partly fermented in the colon.
Another function of food is to supply pleasure. Generally, people enjoy eating and foods deliver stimuli to the
senses (eyes, tongue, nose, ears). Childhood experiences
appear to be very important to what people like in later
years, and also cultural habits have a big influence.
Food is also a very important way to express social relations. Hospitality is expressed by offering food and drinks
to visitors and friends. It is also a way to distinguish oneself
by (not) eating certain foods, usually dictated by religion.
The Future of Food
We have discussed briefly the various aspects of food production, processing, distribution, and consumption. The keywords are food security (is there enough food), food safety
(is the available food safe to consume), and food quality (is
the food of such a quality that it can fulfill the need of the
consumer). Humankind is capable to produce enough food
for the whole world population; in other words, food security can be realized, in principle. In practice, however, this
appears not to be possible due to socioeconomic and political problems. Food safety and food quality are manageable,
again in principle. The question is now how future developments can help in increasing food security, food safety, and
food quality. A rapidly upcoming problem related to food
production is sustainability: Are we able to produce food in
such a way that also future generations are able to fulfill their
needs of food without depleting Mother Earth?
Food and Sustainability
The key sustainability issues in food production are optimal
use of raw materials (with as little waste as possible while
still satisfying consumer demands), efficient usage of water,
energy, packaging materials, and processing aids, and economic efficiency in line with social and cultural values. A
recent concern is that the production of meat and milk is
contributing considerably to environmental problems. One
reason is the emission of gases that are involved in climate
change problems (especially methane). Another reason is
the inefficient conversion of plant proteins in animal feed
into animal proteins for human consumption (Aiking et
al., 2006). This problem is complicated and the simple solution is not to cut down animal production in developing
countries, as the contribution of small-scale animal farming
to livelihood: animals, which are a major source of food,
WWW.CROPS.ORG
S-47
provide fuel and manure to be used as fertilizer in developing countries. In the industrialized world, however, developments of meat alternatives are a possible solution or part
of a solution. It could also be of interest for upcoming countries such as India and China.
Food and Health
The relationship between food, nutrition, and health is
obvious on the one hand, but much is still unknown. First
of all, an individual food cannot be called unhealthy or
healthy, but diets can. In other words, it is the combination
of several foods in a diet that determines what is healthy. In
developing countries, health problems are related mainly
to insufficient energy intake and lack of micronutrients. In
the developed world, the main problem is overconsumption
and too little fiber. Too many calories are taken in with
a minimum of physical exercise, thus leading to obesity
and ultimately to so-called “diseases of civilization” such
as coronary heart disease and diabetes. It is not only food,
by the way, but also lifestyle that determines the incidence
of obesity. Incidentally, obesity is also increasing in certain
populations in the developing world. Obviously, health is
also related to food safety if microorganisms cause diarrhea
with associated problems, and other food intoxications.
THE ROLE OF FOOD TECHNOLOGY
How could food technology help to alleviate some of the
problems that we face? First of all, knowledge of what
causes postharvest losses will help to tackle this problem.
It is estimated that 30 to 40% losses occur in developing
countries and, therefore, measures to avoid this will have a
big effect on food security as well as on food safety (Engstrom and Carlsson-Kanyama, 2004; Kader, 2005). One
of the problems with postharvest spoilage is that microorganisms produce toxins (such as the carcinogenic compound aflatoxin) that are really very dangerous to human
health (Williams et al., 2004). Prevention of this would
definitely help increase food security as well as safety
(Ortiz et al., 2008).
Second, when raw materials are processed into foods,
desired and undesired things happen. Desired effects are
increased digestibility, increased food safety because of
elimination of pathogens, and increased shelf life. Undesired effects are destruction of essential nutrients, and losses
of resources (excessive waste). Because lack of micronutrients is a serious problem in the developing world (Kennedy
et al., 2003), making micronutrients more bioavailable,
possibly by adding them to food, and preventing losses of
these compounds, would make a major contribution to
alleviate inadequate nutrition.
Third, if food processing of local crops can be connected to demand of urban consumers—that is, by aligning food processing to consumers’ wishes—this link
would offer the opportunity to raise income and earn a
S-48
living for the local processors and producers. In doing so,
it is essential that food safety and quality can be guaranteed. Food technology can help in realizing this.
Fourth, institutional barriers to access markets (regionally
and internationally) could be tackled by investing in quality
and safety by technological measures, and to have knowledge
about the products produced so that real safety problems can
be distinguished from trade barriers in disguise.
A very important aspect with all preservation technologies is packaging. It forms the barrier between the
food and its environment, and it can protect the food from
recontamination and other undesired influences from the
environment (such as oxygen). Packaging has therefore a
large effect on food quality as well as on food safety. Food
technology can help in adjusting packaging technology to
what is needed for a particular food.
FOCUSING THE AGRI-FOOD RESEARCH
AGENDA OF THE 21ST CENTURY
Table 2 highlights some issues related to food security and
safety, the link between food and health, and sustainability of agro-ecosystems for both the industrialized and the
developing world. In this regard, agri-food system research
should focus on using better local crops for food production, reducing postharvest losses substantially, optimizing
local processing in such a way that the nutritional value
(especially bioavailability of micronutrients) is improved
as well as the eating quality, linking local production and
processing to urban consumers by delivering safe, nutritious, and good-quality food according to consumers’
demands, improving processing and storage or packaging
to ensure food safety (i.e., absence of pathogens and contaminating chemicals), dealing with institutional barriers
for access to markets, and making more sustainable global
food production systems.
CONCLUSIONS
This background article for the Science Forum 2009 has
provided some scenarios of the future of food for the mid21st century. This article has also briefly addressed the
major developments in food production, especially the
possible role of food technology, and based on this analysis, some issues for shaping a priority agri-food research
agenda have been identified. Next to technological issues,
it should be realized, however, that institutional barriers
can be a major obstacle for the developing world to gain
access to local, regional, and global markets. At the same
time, it is also clear that technology could help to overcome these institutional barriers by making it possible to
produce high-quality, safe, and nutritious food. If it is possible to realize such a development, it should also be possible to reduce poverty by linking urban food demand in
the developing world to local food production.
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
Table 2. Overview of relevant issues for future of food and the role of technology.
Issue
Developing world
Developed world
Is a big issue
Postharvest losses
Postprocessing packaging and storage
No issue anymore: tackled by industrialization of food
production
Pathogenic microorganisms
Contaminating chemicals (pesticides, environmental
contaminants such as dioxin, PCBs†)
Mainly pathogenic microorganisms
Micronutrients
Macronutrients
Soil depletion and use of fertilizers
Water quality
Waste of food
Overconsumption leading to obesity and associated
health problems
Fossil energy use
Water use
Making better use of resources
Waste of food
Food security
Food safety
Food and health
Sustainability
†
PCBs, polychlorinated biphenyls.
Acknowledgments
The authors thank the colleagues who kindly perused an early
version of this manuscript, and Dr. Haruko Okusu (CGIAR
Science Council Secretariat) for her kind editing of the fi nal
version of this manuscript.
References
Agrimonde. 2008. Why a CIRAD-INRA foresight study on
world food and agricultural systems in 2050? Institut National
de la Recherche Agronomique–Centre International de
Recherche Agronomique pour le Développement, Paris.
Agrimonde. 2009. Scenarios and challenges for feeding the
world in 2050. Available at http://www.fao.org/fi leadmin/
templates/wsfs/docs/expert_paper/Inra__Cirad__2009_-_
Agrimonde_Summary_Report__draft_.pdf (verified 21 Dec.
2009). Institut National de la Recherche Agronomique–Centre International de Recherche Agronomique pour le Développement, Paris.
Ahammad, H., and R. Mi. 2005. Land use change modeling in
GTEM: Accounting for forest sinks. ABARE Conf. Pap.
05.13. In Workshop on EMF 22: Climate Change Control
Scenarios, Stanford, CA. 25–27 May 2005. Stanford Univ.,
Stanford, CA.
Aiking, H., J. De Boer, and J. Vereijken. 2006. Sustainable protein
production and consumption: Pigs or peas. Springer, Dordrecht,
Netherlands.
Carpenter, S.R., P.L. Pingali, E.M. Bennett, and M.B. Zurek (ed.)
2005. Ecosystems and human well-being: Scenarios. The
millennium ecosystem assessment. Vol. 2. Island Press, Washington, DC.
Chaumet, J.-M., F. Delpeuch, B. Dorin, G. Ghersi, B. Hubert,
T. Le Cotty, S. Paillard, M. Petit, J.-L. Rastoin, T. Ronzon,
and S. Treyer. 2009. Agriculture and food in the world of
2050—Scenarios and challenges for a sustainable development. Institut National de la Recherche Agronomique–
Centre International de Recherche Agronomique pour le
Développement, Paris.
Collomb, P. 1999. Une voie étroite pour la sécurité alimentaire
d’ici à 2050. Economica, Paris.
Conway, G. 1997. The doubly green revolution—Food for all in
the twenty-fi rst century. Penguin Books, London.
De Jouvenel, H. 2000. A brief methodological guide to scenario
building. Technol. Forecast. Soc. Change 65:37–48.
Eickhout, B., H. van Meijl, and A. Tabeau. 2006 Modeling agricultural trade and food production under different trade policies.
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010
p. 61–76. In A.F. Bouwman et al. (ed.) Integrated modeling
of global environmental change. An overview of IMAGE 2.4.
Netherlands Environ. Assessment Agency (MNP), Bilthoven.
Engstrom, R., and A. Carlsson-Kanyama. 2004. Food losses in
food service institutions. Examples from Sweden. Food Policy
29:203–213.
FAO. 2002. Agriculture mondiale: Horizon 2015/2030. Food and
Agric. Organ. of the United Nations, Rome.
FAO. 2006. World agriculture: Towards 2030/2050. Interim rep.
Food and Agric. Organ. of the United Nations, Rome.
Griffon, M. 2006. Nourrir la planète—Pour une révolution doublement verte. Odile Jacob, Paris.
IAASTD. 2008. Agriculture at a crossroads: Executive summary
of the synthesis report. Int. Assessment of Agric. Knowledge,
Sci., and Technol. for Dev. Island Press, Washington, DC.
Ingram, J.S.I. 2008. Food system concepts for the ESF/COST
Forward Look on European food systems in a changing
world. Work document ESF-COST Forward Look Workshop, Brussels.
Kader, A.A. 2005. Increasing food availability by reducing postharvest losses of fresh produce. Acta Hortic. 682:2169–2175.
Kennedy, G., G. Nantel, and P. Shetty. 2003. The scourge of “hidden hunger”: Global dimensions of micronutrient deficiencies. Food Nutr. Agric. 32:8–16.
Ortiz, R., T. Ban, R. Bandyopadhyay, M. Banziger, D. Bergvinson, K. Hell, B. James, D. Jeffers, P. Lava Kumar, A. Menkir,
J. Murakami, S.N. Nigam, H.D. Upadhyaya, and F. Waliyar. 2008. CGIAR research-for-development program on
mycotoxins. p. 415–424. In J.F. Leslie et al. (ed.) Mycotoxins:
Detection methods, management, public health and agricultural trade. CABI Publishing, Wallingford, UK.
Rosegrant, M.W., X. Cai, and S.A. Cline. 2002. World water and
food to 2025: Dealing with scarcity. Int. Food Policy Res.
Inst., Washington, DC.
Rosegrant, M.W., M. Fernandez, and A. Sinha. 2009. Looking into the future for agriculture and KST. p. 370–376. In
B.D. McIntyre et al. (ed.) IAASTD Global Rep. Island Press,
Washington, DC.
Rosegrant, M.W., M.S. Paisner, S. Meijer, and J. Witcover. 2001.
Global food projections to 2020: Emerging trends and alternative futures. Int. Food Policy Res. Inst., Washington, DC.
Ruben, R., M. van Boekel, A. Van Tilburg, and J. Trienekens (ed.).
2007. Tropical food chains. Governance regimes for quality management. Wageningen Academic Publ., Wageningen, Netherlands.
Thornton, P.K., P.G. Jones, T. Owiyo, R.L. Kruska, M. Herrero,
WWW.CROPS.ORG
S-49
P. Kristjanson, A. Notenbaert, N. Bekele, A. Omolo, with
contributions from V. Orindi, B. Otiende, A. Ochieng, S.
Bhadwal, K. Anantram, S. Nair, V. Kumar, and U. Kulkar.
2006. Mapping climate vulnerability and poverty in Africa—
Report to the Department for International Development.
Int. Livestock Res. Inst., Nairobi.
Thornton, P.K., R.L. Kruska, N. Henninger, P. Kristjanson, R.S.
Reid, F. Atieno, A. Odero, and T. Ndegwa. 2002. Mapping
poverty and livestock in the developing world. Int. Livestock
Res. Inst., Nairobi.
United Nations. 2005. World population prospects: The 2004 revision. United Nations, Population Division of the Dep. of Economic and Social Affairs of the United Nations Secretariat,
S-50
New York.
Van Boekel, M.A.J.S. 1998. Developments in technologies for
food production. p. 87–116. In W.M.F. Jongen and M.T.G.
Meulenberg (ed.) Innovation of food production systems.
Wageningen Press, Wageningen, Netherlands.
Williams, J.H., T.D. Phillips, P.E. Jolly, J.K. Stiles, C.M. Jolly, and
D. Aggarwal. 2004. Human aflatoxicosis in developing countries: A review of toxicology, exposure, potential health consequences, and interventions. Am. J. Clin. Nutr. 80:1106–1122.
World Bank. 2008. World development report 2008: Agriculture
for development. Int. Bank for Reconstruction and Dev.,
Washington, DC.
WWW.CROPS.ORG
CROP SCIENCE, VOL. 50, MARCH– APRIL 2010